Housing.com · Platform health
Left on Read
Buyers said brokers never called. Brokers said buyers never picked up. Everyone was paying for it.
See the screens
ACT 1 Empathize
Two sides, one broken phone line
The feedback came from both directions, and it didn't match.
"The broker never got back to me."
"I call, but buyers don't pick up."
"We're paying for leads. Why aren't they good?"
Everyone had a story, and no one had proof. When a lead went cold, there was no way to tell whose side the conversation had died on.
ACT 2
Why this was worth fixing
Sellers on Housing.com pay for lead packages. Good leads are what make that package worth renewing. When leads feel weak, sellers stop believing in the product, and that's how renewals turn into churn.
The data also pointed to a simple lever: fresh leads convert better. A buyer who just showed interest is far likelier to respond than one who's waited a day. Speed wasn't a courtesy. It was conversion.
Every hour a lead sits untouched, it gets a little colder.
ACT 3 Define
You can't fix what nobody measures
So I defined a new metric: seller responsiveness. Did the broker try to reach the buyer, and how fast?
…know for sure whether a broker reached out to a lead, and make responding fast feel worth it?
First, we had to capture contact attempts. Every new lead already reached the broker three ways: an SMS, an in-app notification, and a new entry on the leads tab of their dashboard. We turned each of those into something we could record.
Turning every touchpoint into a signal
In the SMS, we placed the buyer's phone number behind a tracking link, so tapping it told us an attempt was made. On the platform, chatting via WhatsApp, viewing the contact or using in-app chat were all recorded the same way.
ACT 4 Prototype
Make speed something you can see
Once we could measure it, we showed it back to brokers. Every lead on the leads tab now carries a status based on whether contact was recorded, and how quickly: "Not responded yet!" for leads still waiting, or "Poor response (in 24+ hrs)" when a reply came late. A Not responded filter puts every waiting lead in one place.
We also showed each seller their overall response rate and median response time at the top of their leads tab, so they could see exactly how they were doing.
